A Practical Guide to Building an Agent Harness From Scratch
omarsar0 · x · 2026-09-15
Elvis (omarsar0) shares how to build an agent harness from scratch. He recommends implementing one yourself in TypeScript or Python, starting from the ReAct paper — you can even prompt an agent to read the paper and produce a minimal implementation you can inspect.
Target architecture with three modules:
- LLM module: handles all inference, ideally supporting multiple models (he used OpenRouter when starting); the system prompt can be embedded or kept separate for deeper context-engineering experiments.
- Tools module: recommended to build them as… (truncated).
- Plus the remaining execution/orchestration parts.
Related event: Guide: building a minimal agent harness from scratch(3 posts)→
More from coding & agent
- Claude Mods: Anthropic to ship function hooks-based plugins for Claude Code in weeks — bcherny · 2026-09-15
- OpenAI Agents API enters public beta: managed cloud agents on the Codex harness — craigsdennis · 2026-09-15
- The hidden cost of failed agent runs: one task may burn twice the credits on Codex — ToneAromatic178 · 2026-09-15
- Interview with the Claude Code team on building it while models keep outpacing engineering — EricBuess · 2026-09-15
- Before buying a faster model, check your agent's traces: serial API calls eat latency — gethackteam · 2026-09-15
- Building your own agent harness beats bloated off-the-shelf frameworks on cost, argues researcher — omarsar0 · 2026-09-15